A Computational Study On The Optimization of Transcranial Temporal Interfering Stimulation With High-Definition Electrode Using Unsupervised Neural Network

Abstract

Transcranial temporal interfering stimulation (tTIS) can focally stimulate deep parts of the brain, which are related to specific functions, by using beats at two high AC frequencies that do not affect the human brain. However, it has limitations in terms of calculation time and precision for optimization because of its complexity and non-linearity. We aimed to propose a method using an unsupervised neural network (USNN) for tTIS to optimize quickly the interfering current value of high-definition electrodes, which can finely stimulate the deep part of the brain, and analyze the performance and characteristics of tTIS. A computational study was conducted using 16 realistic head models. This method generated the strongest stimulation on the target, even when targeting deep areas or multi-target stimulation. The tTIS was robust with target depth compared with transcranial alternating current stimulation, and mis-stimulation could be reduced compared with the case of using two-pair inferential stimulation. Optimization of a target could be performed in 3 min. By proposing the USNN for tTIS, we showed that the electrode currents of tTIS can be optimized quickly and accurately, and the possibility of stimulating the deep part of the brain precisely with transcranial electrical stimulation was confirmed.